EP1580987A1 - Solid-state imaging device - Google Patents

Solid-state imaging device Download PDF

Info

Publication number
EP1580987A1
EP1580987A1 EP03768142A EP03768142A EP1580987A1 EP 1580987 A1 EP1580987 A1 EP 1580987A1 EP 03768142 A EP03768142 A EP 03768142A EP 03768142 A EP03768142 A EP 03768142A EP 1580987 A1 EP1580987 A1 EP 1580987A1
Authority
EP
European Patent Office
Prior art keywords
signal
operation period
transistor
detection node
pixel
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
EP03768142A
Other languages
German (de)
French (fr)
Other versions
EP1580987A4 (en
Inventor
Takashi Watanabe
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sharp Corp
Original Assignee
Sharp Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Sharp Corp filed Critical Sharp Corp
Publication of EP1580987A1 publication Critical patent/EP1580987A1/en
Publication of EP1580987A4 publication Critical patent/EP1580987A4/en
Withdrawn legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10F—INORGANIC SEMICONDUCTOR DEVICES SENSITIVE TO INFRARED RADIATION, LIGHT, ELECTROMAGNETIC RADIATION OF SHORTER WAVELENGTH OR CORPUSCULAR RADIATION
    • H10F39/00—Integrated devices, or assemblies of multiple devices, comprising at least one element covered by group H10F30/00, e.g. radiation detectors comprising photodiode arrays
    • H10F39/80—Constructional details of image sensors
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/70—SSIS architectures; Circuits associated therewith
    • H04N25/76—Addressed sensors, e.g. MOS or CMOS sensors
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/70—Circuitry for compensating brightness variation in the scene
    • H04N23/741—Circuitry for compensating brightness variation in the scene by increasing the dynamic range of the image compared to the dynamic range of the electronic image sensors
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/50—Control of the SSIS exposure
    • H04N25/57—Control of the dynamic range
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/50—Control of the SSIS exposure
    • H04N25/57—Control of the dynamic range
    • H04N25/571—Control of the dynamic range involving a non-linear response
    • H04N25/573—Control of the dynamic range involving a non-linear response the logarithmic type
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/60—Noise processing, e.g. detecting, correcting, reducing or removing noise
    • H04N25/65—Noise processing, e.g. detecting, correcting, reducing or removing noise applied to reset noise, e.g. KTC noise related to CMOS structures by techniques other than CDS
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/70—SSIS architectures; Circuits associated therewith
    • H04N25/76—Addressed sensors, e.g. MOS or CMOS sensors
    • H04N25/779—Circuitry for scanning or addressing the pixel array
    • H—ELECTRICITY
    • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10F—INORGANIC SEMICONDUCTOR DEVICES SENSITIVE TO INFRARED RADIATION, LIGHT, ELECTROMAGNETIC RADIATION OF SHORTER WAVELENGTH OR CORPUSCULAR RADIATION
    • H10F39/00—Integrated devices, or assemblies of multiple devices, comprising at least one element covered by group H10F30/00, e.g. radiation detectors comprising photodiode arrays
    • H10F39/10—Integrated devices
    • H10F39/12—Image sensors
    • H10F39/18—Complementary metal-oxide-semiconductor [CMOS] image sensors; Photodiode array image sensors

Definitions

  • the present invention relates to solid-state imaging devices and, more precisely, to a solid-state imaging device that is generally called an image sensor.
  • CMOS Complementary Metal Oxide Semiconductor
  • Tr MOS transistors
  • Fig. 10A shows the construction of one pixel in the case of a PD+3Tr system.
  • the diagram shows a photodiode 1, a detection node 3 and a reset part constructed of a MOS transistor 4 and a drain 5 to which a power voltage V D is applied.
  • An amplification part 6 constructed of a MOS transistor, a pixel selection part 7 constructed of a MOS transistor, a signal line 8, a reset clock ⁇ RS and a pixel selection clock ⁇ S are also shown therein.
  • Fig. 10B shows the operation of Fig. 10A by potential.
  • the photodiode 1 is first reset to the potential V D by reset operation of the reset part 4, and thereafter, a signal charge generated by incident light h v in the photodiode 1 is stored in the detection node 3 in a floating state.
  • a potential V S of the detection node 3 falls from the potential V D due to the storage of electric charge, and the quantity of fall is proportional to the intensity of incident light and a storage period. Therefore, a variation ⁇ V S in the potential V S is proportional to the intensity of incident light in the case of the storage it a certain period.
  • the value is amplified in the amplification part 6 and thereafter selected by the pixel selection part, i.e., switch 7 for read to the signal line 8. Since the signal is proportional to the intensity of incident light in the construction of Fig. 10A, saturation disadvantageously occurs with a sufficiently intense quantity of light, failing in obtaining a wide dynamic range.
  • FIG. 11A is a diagram of a circuit construction of one pixel of the example.
  • a photodiode 1 a detection node 3
  • a logarithmic compression transistor 4 a drain 5 to which a power voltage V D is applied.
  • An amplification part 6, a pixel selection part 7, a signal line 8, a pixel selection clock ⁇ S and the power voltage V D are also shown.
  • Fig. 11B is a diagram showing the operation of the transistor 4 in Fig. 11A by the potential relation.
  • the potential becomes a constant value ⁇ G (H).
  • the transistor 4 operates to perform weak inversion operation, i.e., flow a subthreshold current Isubth. Since the source potential V S changes so that the subthreshold current Isubth becomes equal to the photocurrent Ip, the source potential V S eventually comes to have a value that is proportional to log(Ip), i.e., obtained by logarithmically converting the photocurrent. This makes it possible to achieve responses throughout a very wide range of the quantity of incident light and obtain a very wide dynamic range.
  • the logarithmic conversion type image sensor shown in Figs. 11A and 11B is the device for performing detection in a steady state in which the photocurrent and the subthreshold current are balanced with each other. With a small quantity of incident light, the device cannot use the technique of increasing the amount of signal charge by increasing the storage time as in the storage type image sensor shown in Figs. 10A and 10B. Furthermore,- since a lower limit value Imin of the photocurrent that can be logarithmically converted is restricted by the dark current of the photodiode, an increase in the dark current due to a rise in temperature or the like causes a significant reduction in low-illuminance sensitivity. For the above reasons, the low-illuminance sensitivity of the logarithmic conversion type image sensor is usually inferior to that of the storage type image sensor.
  • Figs. 12A and 12B a system with a single device that exhibits a linear photoelectric conversion characteristic when the optical input is small and a logarithmic photoelectric conversion characteristic when the optical input is large is proposed (refer to, for example, JP H10-90058 A and JP 2000-175108 A).
  • Fig. 12A shows the construction of one pixel including a photodiode 1, a detection node 3, a reset part 4 and a drain 5 to which a power voltage V D is applied as in Fig. 10A.
  • An amplification part 6, a pixel selection part 7, a signal line 8 and a pixel selection clock ⁇ S are also shown.
  • the power voltage V D and a voltage V H that is sufficiently higher than the power voltage V D are alternately applied to the gate V G of the reset part 4 via a switch 9 in a constant cycle.
  • the operation of Fig. 12A is indicated by potentials in Fig. 12B and by timing in Fig. 12C.
  • the voltage V H is first applied to the gate V G of the reset part 4 by the switch 9 in a period T 2 .
  • the potential ⁇ G (V H ) under the gate of the reset part 4 becomes deeper than the power voltage V D , and the potential of the detection node 3 is reset to the power voltage V D .
  • the power voltage V D is applied to the gate V G of the reset part 4 by the switch 9 in a period T 1 .
  • the potential ⁇ G (V D ) under the gate of the reset part 4 becomes shallower than the power voltage V D , and the potential of the detection node 3 enters a floating state.
  • the potential V S of the detection node 3 is reduced from the power supply voltage V D .
  • the quantity of reduction is proportional to the intensity of incident light and the storage period.
  • a variation ⁇ V S 1 of the potential V S of the detection node 3 is proportional to the intensity of incident light.
  • a subthreshold current Isubth flows. Since the potential V S of the detection node 3 is changed by a variation ⁇ V S 2 from the value ⁇ 0 so that the subthreshold current Isubth becomes equal to the photocurrent Ip, and eventually, the value ⁇ V S 2 is proportional to log(Ip). That is, a value obtained by logarithmically converting the photocurrent results.
  • the variation ⁇ V S 1 of the potential V S of the detection node 3 is proportional to the intensity of incident light when V D ⁇ V S > ⁇ 0
  • the variation ⁇ V S 2 of the potential V S of the detection node 3 is proportional to log(Ip) when ⁇ 0 ⁇ V S > ⁇ G (V D ).
  • ⁇ G (V D ) is the potential under the gate of the reset part 4 when the power voltage V D is applied to the gate V G . Therefore, a change in the potential V S of the detection node 3 with respect to the incident light exhibits a linear photoelectric conversion characteristic when the optical input is small and exhibits a logarithmic photoelectric conversion characteristic when the optical input is large as shown in Fig. 12D. As a result, it is possible to provide linear type operation of a high sensitivity at a low illuminance and logarithmic operation of a wide dynamic range at a high illuminance.
  • Figs. 12A, 12B, 12C and 12D has the following problems.
  • the potential value i.e., the boundary between the linear operation and the logarithmic operation ⁇ 0 varies every pixel. Therefore, very large harsh fixed pattern noises are generated without modification in the logarithmic operation region.
  • k represents the Boltzman's constant
  • T represents the absolute temperature
  • q represents the amount of electronic charge.
  • An object of the present invention is to solve the various problems and provide a solid-state imaging device capable of achieving both of a wide dynamic range and a high low-illuminance sensitivity.
  • a solid-state imaging device in which a photodiode and a first transistor are provided in series between a ground and a drain in each pixel, and a signal corresponding to a current or an electric charge generated in the photodiode according to an optical input is outputted from a detection node located between the photodiode and the first transistor, comprising:
  • the "first level” and the “second level” should properly be set to levels at which the potentials of the signal charge become deeper and shallower, respectively, just under the gate of the first transistor.
  • the "first level” and the “second level” correspond to High level and Low level, respectively.
  • the logarithmic operation period during which the logarithmically converted photoelectric conversion signal is obtained by setting the gate voltage of the first transistor to the first level and the linear operation period during which the linear type photoelectric conversion signal is obtained by setting the gate voltage of the first transistor to the second level are alternately repeated under the control of the control part.
  • the logarithmically converted photoelectric conversion signal can be obtained at the detection node. Therefore, by taking out the signal from the detection node and transferring the signal, a logarithmic signal with a wide dynamic range is outputted.
  • a linear photoelectric conversion signal can be obtained at the detection node. Therefore, by taking out the signal from the detection node and transferring the signal, a linear signal of a high sensitivity under a low illuminance is outputted. Therefore, according to the solid-state imaging device, both of the wide dynamic range and the low-illuminance high sensitivity can be achieved.
  • the photodiode and the detection node are connected to each other. That is, one terminal of the photodiode and the detection node may be short-circuited to each other.
  • a second transistor is connected between the photodiode and the detection node.
  • the second transistor is connected between the photodiode and the detection node in the solid-state imaging device of the embodiment, it is possible to reduce the capacitance of the detection node and enhance the electric charge voltage conversion efficiency during the linear operation period.
  • the photodiode has a buried-channel structure.
  • the photodiode Since the photodiode has the buried-channel structure in the solid-state imaging device of the embodiment, it becomes possible to largely reduce the dark current occurring in the photodiode. Therefore, it becomes possible to extend the lower limit of the photocurrent that can be logarithmically converted during the logarithmic operation period. Moreover, dark current noises can also be reduced during the linear operation period.
  • control part executes control so as to
  • the logarithmic operation period and the linear operation period are alternately repeated every frame.
  • the photoelectrically converted electric charge starts to be stored in the detection node.
  • a largest amount of electric charge is stored in the detection node. If the electric charge is read as a linear signal, a high-sensitivity output is obtained.
  • the electric charge can be read as a logarithmic signal from the detection node during the logarithmic operation period after a lapse of a certain period after the transition to the logarithmic operation period.
  • the solid-state imaging device of one embodiment comprises:
  • the signal read from the detection node of each pixel within the logarithmic operation period under the condition that light is irradiated with a certain uniform intensity to each pixel is recorded in the first frame memory. Subsequently, the signal recorded in the first frame memory is subtracted from the signal read in an arbitrary frame in correspondence with each pixel. Therefore, variation in the characteristics of the pixels and, in particular, the characteristic variation (referred to as an "offset variation") attributed to the variation in the threshold values of the transistors can be canceled. Therefore, an image with reduced amount of fixed pattern noises can be obtained.
  • the subtraction part forms an output by subtracting the signal recorded in the first frame memory from a signal read in the logarithmic operation period in association with each pixel.
  • an image of a wide dynamic range and reduced amount of fixed pattern noises can be obtained.
  • the solid-state imaging device of one embodiment comprises:
  • the signal read from the detection node immediately before the transition from the logarithmic operation period to the linear operation period under the condition that light is irradiated with a certain uniform intensity to each pixel is recorded in the second frame memory. Subsequently, the signal recorded in the second frame memory is subtracted from the signal read from the detection node immediately before the transition from the linear operation period to the logarithmic operation period in association with each pixel. Therefore, it becomes possible to take out only the net linear signal component. Furthermore, a difference between the first and last signals of the storage of the signal charge is taken in this case, and therefore, a high-sensitivity image from which the reset noises are completely removed can be obtained.
  • Fig. 5 is a diagram that shows a two-dimensional image sensor 10 of one embodiment of the present invention with a circuit construction of 2 x 2 pixels.
  • a reference numeral 11 denotes a pixel of a circuit construction described later
  • 12 denotes a line for a reset clock ⁇ R applied to a first transistor
  • 13 denotes a line for a pixel selection clock ⁇ S
  • 14 denotes a line for a signal Vsig
  • 15 denotes a power voltage V D .
  • the reset clock ⁇ R and the pixel selection clock ⁇ S are successively outputted in rows from a reset scan circuit 16 and a vertical read scan circuit 17. The rows of pixels are successively scanned in a vertical direction.
  • the signals Vsig read in rows from the pixels are successively read in the horizontal direction to a horizontal signal line 18 by a signal from a horizontal read scan circuit 19.
  • a signal from the horizontal signal line 18 is outputted as an output signal OS via an amplifier circuit 20.
  • the entire operation of the two-dimensional image sensor 10 is controlled by a CPU (Central Processing Unit) 90 as one example of a control part:
  • Fig. 1A illustrates a circuit construction example of each of the pixels 11 shown in Fig. 5.
  • a photodiode 1 a detection node 3, a first transistor 4 and a drain 5 to which a power voltage V D is applied.
  • an amplification part 6 constructed of a MOS transistor
  • a pixel selection part 7 constructed of a MOS transistor
  • a signal line 8 a reset clock ⁇ R and a pixel selection clock ⁇ S .
  • the photodiode 1 and the first transistor 4 are provided in series between the ground and the drain 5.
  • the logarithmic operation and the linear operation are performed as follows by using the pixel 11.
  • Fig. 2A shows potential relations when the logarithmic operation is performed by using the pixel of Fig. 1A.
  • the gate of the first transistor 4 is maintained at the DC (Direct Current) level, and the potential has a constant value ⁇ G (H).
  • the source potential V S of the first transistor 4 becomes deeper than the constant value ⁇ G (H)
  • K 1 and K 2 are constants. This makes it possible to achieve responses throughout a very wide range of the quantity of incident light and obtain a very wide dynamic range.
  • Fig. 3A shows potential relations when the linear operation is performed by using the pixel of Fig. 1A.
  • a pulse ⁇ R is applied to the gate of the first transistor 4.
  • the gate of the first transistor 4 has been maintained at High level for a sufficiently long period before the start of signal storage, and the source potential V S comes to have a value V S (log) at which the photocurrent Ip and the subthreshold current Isubth become equal to each other.
  • V S log
  • the gate of the first transistor 4 changes to Low level, and the signal storage starts. After a lapse of one frame period, the source potential is reduced to V S (lin).
  • ⁇ T represents the storage time
  • C 1 represents the capacitance of the detection node 3 shown in Fig. 1A.
  • Fig. 1B illustrates the other Circuit construction example of each of the pixels 11 shown in Fig. 5 different from that of Fig. 1A.
  • the circuit construction of Fig. 1B differs from the one of Fig. 1A in that a second transistor 2 is inserted between the photodiode 1 and the detection node 3.
  • a DC potential ⁇ T is applied to the gate of the transistor 2.
  • Figs. 2B shows potential relations when the logarithmic operation is performed by using the pixel of Fig. 1B.
  • a photocurrent Ip is generated in the photodiode 1
  • a current corresponding to the photocurrent Ip flows through the second transistor 2 in the steady state since the potential ⁇ T at the gate of the second transistor 2 is a DC potential, and the potential of the photodiode 1 is maintained at a constant value.
  • the potential ⁇ R of the gate of the first transistor 4 is a DC potential
  • a subthreshold current Isubth flows.
  • the potential V S of the detection node 3 changes so that the subthreshold current Isubth becomes equal to the photocurrent Ip, the potential V S of the detection node 3 comes to have a value V S (log) obtained by logarithmically converting the photocurrent Ip according to Equation (1). This makes it possible to achieve responses throughout a very wide range of the quantity of incident light and obtain a very wide dynamic range.
  • Fig. 3B shows potential relations when the linear operation is performed by using the pixel of Fig. 1B.
  • the photocurrent Ip is generated in the photodiode 1
  • a current corresponding to the photocurrent Ip flows through the gate in the steady state since the potential ⁇ T of the gate of the second transistor 2 is a DC potential, and the potential of the photodiode 1 is maintained at a constant value.
  • the pulse ⁇ R is applied to the gate of the first transistor 4.
  • the first transistor 4 has been maintained at High level for a sufficiently long period before the signal storage starts, and the source voltage comes to have a value V S (log) at which the photocurrent Ip and the subthreshold current becomes equal to each other.
  • ⁇ V S (Ip ⁇ T)/C 2 which expressed a value obtained by linearly converting the photocurrent Ip.
  • ⁇ T represents the storage time
  • C 2 represents the capacitance of the detection node 3 shown in Fig. 1B.
  • C 1 (capacitance of photodiode 1 + gate capacitance of transistor 6 + stray capacitance of wiring and so on)
  • C 2 (gate capacitance of transistor 6 + stray capacitance of wiring and so on)
  • the area of the detection node 3 of Fig. 3B can be made sufficiently smaller than the area of the photodiode 1 of Fig. 3A, and therefore, C 1 > C 2 . That is, it becomes possible to obtain in the case of Fig. 3B a signal voltage ⁇ V S higher than in the case of Fig. 3A with the same amount of signal charge I P ⁇ T.
  • Fig. 4A schematically shows the sectional structure of the pixel of Fig. 1A fabricated in a semiconductor substrate.
  • Figs. 4B and 4C schematically show the sectional structure of the pixel of Fig. 1B fabricated in a semiconductor substrate.
  • a reference numeral 101 denotes a semiconductor substrate
  • 102 denotes a pixel isolation region
  • 103 denotes a cathode of the photodiode 1 (see Figs. 1A and 1B)
  • 104 denotes the drain 5
  • 111 denotes the first transistor 4.
  • a reference numeral 105 denotes an isolated detection node, which is separated from the cathode 103 of the photodiode 1 via the second transistor 112.
  • the photodiode 1 has a simple PN junction structure and is formed concurrently with the drain 104 in Figs. 4A and 4B
  • the photodiode has a buried-channel structure and is formed separately from the drain in Fig. 4C. That is, a signal charge storage layer 106 is formed on the substrate side, and a heavily doped pinning layer 107 is formed on the surface side.
  • a buried-channel structure photodiode is allowed to have a largely reduced dark current in comparison with the simple PN junction structure. This makes it possible to extend the lower limit Imin of the photocurrent capable of being logarithmically converted during the logarithmic operation. Moreover, dark current noises can be reduced also in the linear operation.
  • Fig. 6 shows the operation timing of the two-dimensional image sensor 10 shown in Fig. 5.
  • ⁇ R (1) and ⁇ R (2) represent the reset clocks of the first row and the second row
  • ⁇ S (1) and ⁇ S (2) represent the pixel selection clocks of the first row and the second row
  • OS represents an output signal.
  • 1H represents one horizontal scan period
  • 1V represents one frame period.
  • a linear type photoelectric conversion signal Lin(1) is outputted as the output signal OS.
  • Figs. 7A and Fig. 7B are graphs showing a logarithmically converted photoelectric conversion signal Vs(log) and a linear type photoelectric conversion signal Vs(lin) obtained as the potentials of the detection node 3 of the pixel with the logarithm log(Ip) of the intensity of incident light represented by the horizontal axis (since the change in the potential Vs of the detection node 3 with respect to an increase in the incident light is in the negative direction in Figs. 1A and 1B, curves are inversely shown in Figs. 7A and 7B for the sake of convenience).
  • the logarithmically converted photoelectric conversion signal Vs(log) does not depend on the read period, or the length of one frame period (properly referred to as an "1V period") in the case of Fig. 6.
  • the linear type photoelectric conversion signal Vs(lin) has a signal storage period of 1V, and therefore, the output is increased as the period 1V is prolonged.
  • Fig. 7A corresponds to the case where the period 1V is long
  • Fig. 7B corresponds to the case where the period 1V is short.
  • the value of Vs(log) with respect to the intensity of incident light has a response lower limit value Imin limited by the dark current and has an upper limit value being extremely higher than the linear type photoelectric conversion signal Vs(lin).
  • the value of ⁇ Vs becomes sufficiently greater than Vs(log), and Vs(lin) comes to have a linear graphic curve almost identical to ⁇ Vs.
  • the value of ⁇ Vs becomes lower than Vs(log) on the low incident light intensity side, and Vs(lin) comes to have a nonlinear graphic curve considerably changed from ⁇ Vs.
  • Fig. 7C is a graph showing the output signal of the image sensor of the present invention with the logarithm log(Ip) of the intensity of incident light represented by the horizontal axis.
  • the solid line indicates a mean value ⁇ OS> of all the pixels
  • the dashed line indicates a value OSij of a specified pixel (pixel having an address of i-th row and j-th column).
  • the response of each pixel accompanies a specific offset variation ⁇ Vij due to the variations of the thresholds of the transistors 4 and 6, and the value of ⁇ Vij varies every pixel. Therefore, when the response of each pixel is directly used as an image signal, the image quality is largely impaired by the harsh fixed pattern noises of ⁇ Vij.
  • Fig. 8A illustrates a circuit construction 30 for solving the problems concerning the fixed pattern noises.
  • an analog signal from the image sensor 31 (identical to the image sensor 10 shown in Fig. 5) of the present invention is converted into a digital signal by an AD converter 33.
  • the signal from the AD converter 33 is diverged so that one branch is guided directly to a difference circuit 37 as one example of a subtraction part and the other branch is guided to the difference circuit 37 via a frame memory 34 served as a first frame memory.
  • a logarithmically converted photoelectric conversion signal is outputted from the image sensor 31 under the condition that light is irradiated to each pixel with a uniform intensity (denoted by Ip1 in Fig.
  • the signal is recorded in the frame memory 34 every pixel.
  • the offset variation ⁇ Vij of each pixel is recorded in the frame memory 34.
  • the difference circuit 37 subtracts the signal recorded in the frame memory 34 from the signal read in an arbitrary frame under the subject imaging condition in association with each pixel.
  • the offset variation ⁇ Vij is canceled in all the frame signals, i.e., the logarithmically converted photoelectric conversion signal OS(log) and the linear type photoelectric conversion signal OS(lin), and an image signal free from the fixed harsh noises can be obtained.
  • the linear type photoelectric conversion signal OS(lin) in Fig. 8A has a characteristic that the logarithmic characteristic and the linear characteristic are added to each other although the offset variation ⁇ Vij is canceled.
  • a problem occurs due to a nonlinear characteristic when the linear characteristic value becomes lower than the logarithmic characteristic value as in Fig. 7B.
  • Fig. 8B illustrates another circuit construction 40 for solving the problem concerning the linear characteristic.
  • an analog signal from the image sensor 31 of the present invention is converted into a digital signal by the AD converter 33.
  • the signal from the AD converter 33 is diverged into three branches, and the first branch is directly guided to the difference circuit 37 as one example of the subtraction part.
  • the second branch is guided to the difference circuit 37 via a frame memory 34 as a first frame memory and a changeover switch 36.
  • the third branch is guided to the difference circuit 37 via a frame memory 35 as a second frame memory and the changeover switch 36.
  • a logarithmically converted photoelectric conversion signal from the image sensor 31 is recorded every pixel in the frame memory 34 under the condition that light is applied with a certain uniform intensity to each pixel as in the case of Fig. 8A.
  • the offset variation ⁇ Vij of each pixel is recorded in the frame memory 34.
  • a logarithmic signal read immediately before the reset gate voltage changes from High level to Low level under the subject imaging condition is recorded in a rewrite manner in the frame memory 35. Since the changeover switch 36 is connected to the frame memory 35 side during linear signal read, the signal recorded in the frame memory 35 is subtracted from the linear signal read immediately before the reset gate voltage changes next from Low level to High level in correspondence with each pixel.
  • the changeover switch 36 is connected to the frame memory 34 side during the logarithmic signal read, and therefore, the difference circuit 37 subtracts the signal recorded in the frame memory 34 from the logarithmic signal in correspondence with each pixel.
  • the offset variation ⁇ Vij is canceled in the logarithmically converted photoelectric conversion signal OS(log), and an image signal free from fixed harsh noises can be obtained.
  • Fig. 9A is the same as the case of Fig. 6, and Fig. 9B shows a case where the linear operation period of one frame and the logarithmic operation period of two frames are alternately repeated.
  • Fig. 9C shows a case where the linear operation period of one frame and the logarithmic operation period of three frames are alternately repeated.
  • other combinations are, of course, possible.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Nonlinear Science (AREA)
  • Solid State Image Pick-Up Elements (AREA)
  • Transforming Light Signals Into Electric Signals (AREA)

Abstract

There is provided a solid-state imaging device capable of achieving both of a wide dynamic range and a high low-illuminance sensitivity. A photodiode and a first transistor are provided in series between the ground and a drain of each of pixels, and a signal corresponding to a current or electric charge generated in the photodiode in accordance with an optical input is outputted from a detection node located between the photodiode and the first transistor. A control part executes control to alternately repeat a logarithmic operation period during which a photoelectric conversion signal logarithmically converted by setting a gate voltage R of the first transistor to a first level is obtained and a linear operation period during which a linear type photoelectric conversion signal is obtained by setting the gate voltage R of the first transistor to a second level.

Description

TECHNICAL FIELD
The present invention relates to solid-state imaging devices and, more precisely, to a solid-state imaging device that is generally called an image sensor.
BACKGROUND ART
Recently, amplification type solid-state imaging devices in which the pixels are each provided with an amplification capability and read is performed by a scan circuit and, in particular, CMOS (Complementary Metal Oxide Semiconductor) type image sensors of which the pixels, peripheral drive circuits and signal processing circuits are of CMOS type, are widely used. In the CMOS type image sensor, it is necessary to form a photoelectric conversion part, an amplification part, a pixel selection part and so on in one pixel, and several MOS transistors (hereinafter sometimes abbreviated to Tr) are normally employed besides the photoelectric conversion part constructed of a photodiode (hereinafter sometimes abbreviated to PD).
Fig. 10A shows the construction of one pixel in the case of a PD+3Tr system. The diagram shows a photodiode 1, a detection node 3 and a reset part constructed of a MOS transistor 4 and a drain 5 to which a power voltage VD is applied. An amplification part 6 constructed of a MOS transistor, a pixel selection part 7 constructed of a MOS transistor, a signal line 8, a reset clock RS and a pixel selection clock S are also shown therein. Fig. 10B shows the operation of Fig. 10A by potential.
In Figs. 10A and 10B, the photodiode 1 is first reset to the potential VD by reset operation of the reset part 4, and thereafter, a signal charge generated by incident light hv in the photodiode 1 is stored in the detection node 3 in a floating state. A potential VS of the detection node 3 falls from the potential VD due to the storage of electric charge, and the quantity of fall is proportional to the intensity of incident light and a storage period. Therefore, a variation ΔVS in the potential VS is proportional to the intensity of incident light in the case of the storage it a certain period. The value is amplified in the amplification part 6 and thereafter selected by the pixel selection part, i.e., switch 7 for read to the signal line 8. Since the signal is proportional to the intensity of incident light in the construction of Fig. 10A, saturation disadvantageously occurs with a sufficiently intense quantity of light, failing in obtaining a wide dynamic range.
Accordingly, as shown in Figs. 11A and 11B, a system, in which a photocurrent is read by logarithmic compression in order to obtain a wide dynamic range of incident light, is proposed. Fig. 11A is a diagram of a circuit construction of one pixel of the example. Although the case of an n-channel type is described below, the same argument can be similarly applied by inverting the polarity in the case of a p-channel type. The diagram shows a photodiode 1, a detection node 3, a logarithmic compression transistor 4 and a drain 5 to which a power voltage VD is applied. An amplification part 6, a pixel selection part 7, a signal line 8, a pixel selection clock S and the power voltage VD are also shown. What is significantly different from the case of Fig. 10A is the arrangement that the DC (Direct Current) voltage VD is applied to the gate of the transistor 4 and the logarithmic compression is performed instead of the reset operation. The operation is described below. Fig. 11B is a diagram showing the operation of the transistor 4 in Fig. 11A by the potential relation.
As shown in Fig. 11A, since the gate voltage of the transistor 4 is fixed to the DC potential VD, the potential becomes a constant value ψG(H). When the source potential VS of the transistor 4 becomes deeper than the constant value ψG(H), the transistor 4 operates to perform weak inversion operation, i.e., flow a subthreshold current Isubth. Since the source potential VS changes so that the subthreshold current Isubth becomes equal to the photocurrent Ip, the source potential VS eventually comes to have a value that is proportional to log(Ip), i.e., obtained by logarithmically converting the photocurrent. This makes it possible to achieve responses throughout a very wide range of the quantity of incident light and obtain a very wide dynamic range.
The logarithmic conversion type image sensor shown in Figs. 11A and 11B is the device for performing detection in a steady state in which the photocurrent and the subthreshold current are balanced with each other. With a small quantity of incident light, the device cannot use the technique of increasing the amount of signal charge by increasing the storage time as in the storage type image sensor shown in Figs. 10A and 10B. Furthermore,- since a lower limit value Imin of the photocurrent that can be logarithmically converted is restricted by the dark current of the photodiode, an increase in the dark current due to a rise in temperature or the like causes a significant reduction in low-illuminance sensitivity. For the above reasons, the low-illuminance sensitivity of the logarithmic conversion type image sensor is usually inferior to that of the storage type image sensor.
Accordingly, as shown in Figs. 12A and 12B, a system with a single device that exhibits a linear photoelectric conversion characteristic when the optical input is small and a logarithmic photoelectric conversion characteristic when the optical input is large is proposed (refer to, for example, JP H10-90058 A and JP 2000-175108 A). Fig. 12A shows the construction of one pixel including a photodiode 1, a detection node 3, a reset part 4 and a drain 5 to which a power voltage VD is applied as in Fig. 10A. An amplification part 6, a pixel selection part 7, a signal line 8 and a pixel selection clock S are also shown. The power voltage VD and a voltage VH that is sufficiently higher than the power voltage VD are alternately applied to the gate VG of the reset part 4 via a switch 9 in a constant cycle. The operation of Fig. 12A is indicated by potentials in Fig. 12B and by timing in Fig. 12C. In Figs. 12A, 12B and 12C, the voltage VH is first applied to the gate VG of the reset part 4 by the switch 9 in a period T2. At this time, the potential ψG (VH) under the gate of the reset part 4 becomes deeper than the power voltage VD, and the potential of the detection node 3 is reset to the power voltage VD. Next, the power voltage VD is applied to the gate VG of the reset part 4 by the switch 9 in a period T1. At this time, the potential ψG (VD) under the gate of the reset part 4 becomes shallower than the power voltage VD, and the potential of the detection node 3 enters a floating state. When a signal charge is generated by incident light hv in the photodiode 1, the signal charge is stored in the detection node 3. In accordance with the storage of the signal charge, the potential VS of the detection node 3 is reduced from the power supply voltage VD. The quantity of reduction is proportional to the intensity of incident light and the storage period. Therefore, in the storage of a certain period, a variation ΔVS1 of the potential VS of the detection node 3 is proportional to the intensity of incident light. When the potential VS of the detection node 3 is reduced to a certain voltage value ψ0, weak inversion operation occurs, i.e., a subthreshold current Isubth flows. Since the potential VS of the detection node 3 is changed by a variation ΔVS2 from the value ψ0 so that the subthreshold current Isubth becomes equal to the photocurrent Ip, and eventually, the value ΔVS2 is proportional to log(Ip). That is, a value obtained by logarithmically converting the photocurrent results.
According to the above, the variation ΔVS1 of the potential VS of the detection node 3 is proportional to the intensity of incident light when VD ≥ VS > ψ0, and the variation ΔVS2 of the potential VS of the detection node 3 is proportional to log(Ip) when ψ0 ≥ VS > ψG (VD). In this case, ψG(VD) is the potential under the gate of the reset part 4 when the power voltage VD is applied to the gate VG. Therefore, a change in the potential VS of the detection node 3 with respect to the incident light exhibits a linear photoelectric conversion characteristic when the optical input is small and exhibits a logarithmic photoelectric conversion characteristic when the optical input is large as shown in Fig. 12D. As a result, it is possible to provide linear type operation of a high sensitivity at a low illuminance and logarithmic operation of a wide dynamic range at a high illuminance.
However, the system of Figs. 12A, 12B, 12C and 12D has the following problems. First, the potential value, i.e., the boundary between the linear operation and the logarithmic operation ψ0 varies every pixel. Therefore, very large harsh fixed pattern noises are generated without modification in the logarithmic operation region. Next, since the detection node 3 (assumed to have a capacitance C1) is reset every time in the photodetection operation, so-called kTC noises (thermal noises) expressed by electron count as: Δn = (kTC1) 1/2/q occur and become random noises. In the equation, k represents the Boltzman's constant, T represents the absolute temperature and q represents the amount of electronic charge. These fixed pattern noises and random noises largely deteriorate the image quality.
DISCLOSURE OF THE INVENTION
An object of the present invention is to solve the various problems and provide a solid-state imaging device capable of achieving both of a wide dynamic range and a high low-illuminance sensitivity.
In order to solve the problems, according to the present invention, there is provided a solid-state imaging device in which a photodiode and a first transistor are provided in series between a ground and a drain in each pixel, and a signal corresponding to a current or an electric charge generated in the photodiode according to an optical input is outputted from a detection node located between the photodiode and the first transistor, comprising:
  • a control part that executes control to alternately repeat a logarithmic operation period during which a photoelectric conversion signal logarithmically converted by setting a gate voltage of the first transistor to a first level is obtained and a linear operation period during which a linear photoelectric conversion signal is obtained by setting the gate voltage of the first transistor to a second level.
    • The "first level" and the "second level" should properly be set to levels at which the potentials of the signal charge become deeper and shallower, respectively, just under the gate of the first transistor. For example, in the case of an n-channel type solid-state imaging device, the "first level" and the "second level" correspond to High level and Low level, respectively.
    In the solid-state imaging device of the present invention, the logarithmic operation period during which the logarithmically converted photoelectric conversion signal is obtained by setting the gate voltage of the first transistor to the first level and the linear operation period during which the linear type photoelectric conversion signal is obtained by setting the gate voltage of the first transistor to the second level are alternately repeated under the control of the control part.
    During the logarithmic operation period, the logarithmically converted photoelectric conversion signal can be obtained at the detection node. Therefore, by taking out the signal from the detection node and transferring the signal, a logarithmic signal with a wide dynamic range is outputted. On the other hand, during the linear operation period, a linear photoelectric conversion signal can be obtained at the detection node. Therefore, by taking out the signal from the detection node and transferring the signal, a linear signal of a high sensitivity under a low illuminance is outputted. Therefore, according to the solid-state imaging device, both of the wide dynamic range and the low-illuminance high sensitivity can be achieved.
    In the solid-state imaging device of one embodiment, the photodiode and the detection node are connected to each other. That is, one terminal of the photodiode and the detection node may be short-circuited to each other.
    In one embodiment, a second transistor is connected between the photodiode and the detection node.
    Since the second transistor is connected between the photodiode and the detection node in the solid-state imaging device of the embodiment, it is possible to reduce the capacitance of the detection node and enhance the electric charge voltage conversion efficiency during the linear operation period.
    In one embodiment, the photodiode has a buried-channel structure.
    Since the photodiode has the buried-channel structure in the solid-state imaging device of the embodiment, it becomes possible to largely reduce the dark current occurring in the photodiode. Therefore, it becomes possible to extend the lower limit of the photocurrent that can be logarithmically converted during the logarithmic operation period. Moreover, dark current noises can also be reduced during the linear operation period.
    In one embodiment, the control part executes control so as to
  • alternately repeat the logarithmic operation period and the linear operation period every frame,
  • read a potential of the detection node as a linear type signal immediately before a transition from the linear operation period to the logarithmic operation period, and
  • read the potential of the detection node as a logarithmic signal in the logarithmic operation period after a lapse of a certain period after the transition to the logarithmic operation period.
    • In the solid-state imaging device of the embodiment, the logarithmic operation period and the linear operation period are alternately repeated every frame. After the transition to the linear operation period, i.e., after the gate voltage of the first transistor changes from the first level to the second level, the photoelectrically converted electric charge starts to be stored in the detection node. Next, immediately before the transition from the linear operation period to the logarithmic operation period, i.e., immediately before the gate voltage of the first transistor changes from the second level to the first level, a largest amount of electric charge is stored in the detection node. If the electric charge is read as a linear signal, a high-sensitivity output is obtained. Moreover, it can be considered that a steady state in which the photocurrent and the subthreshold current are balanced with each other is achieved after a lapse of a certain period after the gate voltage of the first transistor has changed from the second level to the first level. Therefore, the electric charge can be read as a logarithmic signal from the detection node during the logarithmic operation period after a lapse of a certain period after the transition to the logarithmic operation period.
    The solid-state imaging device of one embodiment comprises:
  • a first frame memory which stores a signal read from the detection node of each pixel in the logarithmic operation period under a condition that light is irradiated to each pixel with a certain uniform intensity; and
  • a subtraction part which subsequently forms an output by subtracting the signal recorded in the first frame memory from a signal read in an arbitrary frame in association with each pixel.
    • In the solid-state imaging device of the embodiment, the signal read from the detection node of each pixel within the logarithmic operation period under the condition that light is irradiated with a certain uniform intensity to each pixel is recorded in the first frame memory. Subsequently, the signal recorded in the first frame memory is subtracted from the signal read in an arbitrary frame in correspondence with each pixel. Therefore, variation in the characteristics of the pixels and, in particular, the characteristic variation (referred to as an "offset variation") attributed to the variation in the threshold values of the transistors can be canceled. Therefore, an image with reduced amount of fixed pattern noises can be obtained.
    In one embodiment, the subtraction part forms an output by subtracting the signal recorded in the first frame memory from a signal read in the logarithmic operation period in association with each pixel.
    In the solid-state imaging device of the embodiment, an image of a wide dynamic range and reduced amount of fixed pattern noises can be obtained.
    The solid-state imaging device of one embodiment comprises:
  • a second frame memory which records the signal read from the detection node every time immediately before the transition from the logarithmic operation period to the linear operation period under a subject imaging condition; and
  • a subtraction part which subtracts the signal recorded in the second frame memory from the signal read from the detection node immediately before the transition from the linear operation period to the logarithmic operation period in association with each pixel.
    • In the solid-state imaging device of the embodiment, the signal read from the detection node immediately before the transition from the logarithmic operation period to the linear operation period under the condition that light is irradiated with a certain uniform intensity to each pixel is recorded in the second frame memory. Subsequently, the signal recorded in the second frame memory is subtracted from the signal read from the detection node immediately before the transition from the linear operation period to the logarithmic operation period in association with each pixel. Therefore, it becomes possible to take out only the net linear signal component. Furthermore, a difference between the first and last signals of the storage of the signal charge is taken in this case, and therefore, a high-sensitivity image from which the reset noises are completely removed can be obtained.
    BRIEF DESCRIPTION OF THE DRAWINGS
  • Figs. 1A and 1B are diagrams showing circuit constructions of pixels used for a two-dimensional image sensor of one embodiment of a solid-state imaging device of the present invention;
  • Figs. 2A and 2B are diagrams showing potential distributions in logarithmic operation of the pixels shown in Figs. 1A and 1B;
  • Figs. 3A and 3B are diagrams showing potential distributions in linear operation of the pixels shown in Figs. 1A and 1B;
  • Figs. 4A, 4B and 4C are views showing sectional structures when the pixels shown in Figs. 1A and 1B are fabricated in a semiconductor substrate;
  • Fig. 5 is a diagram showing the circuit construction of a two-dimensional image sensor of one embodiment of a solid-state imaging device of the present invention;
  • Fig. 6 is a chart showing the operation timing of the two-dimensional image sensor shown in Fig. 5;
  • Figs. 7A, 7B and 7C are graphs showing a logarithmic signal and a linear signal obtained by the present invention in relation to the intensity of incident light;
  • Figs. 8A and 8B are diagrams showing systems for performing image signal processing with the two-dimensional image sensor of the present invention;
  • Figs. 9A, 9B and 9C are charts showing other examples of the operation timing in the two-dimensional image sensor of the present invention;
  • Figs. 10A and 10B are diagrams for explaining the operation of the pixel of a conventional linear conversion type solid-state imaging device;
  • Figs. 11A and 11B are diagrams for explaining the operation of the pixel of a conventional logarithmic conversion type solid-state imaging device; and
  • Figs. 12A, 12B, 12C and 12D are diagrams for explaining the operation of the pixel of a conventional solid-state imaging device in which the linear conversion characteristic and the logarithmic conversion characteristic are added to each other.
    • BEST MODE FOR CARRYING OUT THE INVENTION
    The present invention will be described in detail below by the embodiments shown in the drawings.
    Fig. 5 is a diagram that shows a two-dimensional image sensor 10 of one embodiment of the present invention with a circuit construction of 2 x 2 pixels. In the two-dimensional image sensor 10, a reference numeral 11 denotes a pixel of a circuit construction described later, 12 denotes a line for a reset clock R applied to a first transistor, 13 denotes a line for a pixel selection clock S, 14 denotes a line for a signal Vsig and 15 denotes a power voltage VD. The reset clock R and the pixel selection clock S are successively outputted in rows from a reset scan circuit 16 and a vertical read scan circuit 17. The rows of pixels are successively scanned in a vertical direction. The signals Vsig read in rows from the pixels are successively read in the horizontal direction to a horizontal signal line 18 by a signal from a horizontal read scan circuit 19. A signal from the horizontal signal line 18 is outputted as an output signal OS via an amplifier circuit 20. The entire operation of the two-dimensional image sensor 10 is controlled by a CPU (Central Processing Unit) 90 as one example of a control part:
    Fig. 1A illustrates a circuit construction example of each of the pixels 11 shown in Fig. 5. In the figure are shown a photodiode 1, a detection node 3, a first transistor 4 and a drain 5 to which a power voltage VD is applied. Further shown are an amplification part 6 constructed of a MOS transistor, a pixel selection part 7 constructed of a MOS transistor, a signal line 8, a reset clock R and a pixel selection clock S. The photodiode 1 and the first transistor 4 are provided in series between the ground and the drain 5. The logarithmic operation and the linear operation are performed as follows by using the pixel 11.
    Fig. 2A shows potential relations when the logarithmic operation is performed by using the pixel of Fig. 1A. In this case, the gate of the first transistor 4 is maintained at the DC (Direct Current) level, and the potential has a constant value ψG(H). When the source potential VS of the first transistor 4 becomes deeper than the constant value ψG(H), the first transistor 4 enters weak inversion operation, and a subthreshold current Isubth flows. Since the source potential VS changes so that the subthreshold current Isubth becomes equal to the photocurrent Ip, VS = K1·log(Ip) + K2 holds, and the source potential VS comes to have a value VS(log) obtained by logarithmically converting the photocurrent Ip. In this case, K1 and K2 are constants. This makes it possible to achieve responses throughout a very wide range of the quantity of incident light and obtain a very wide dynamic range.
    Fig. 3A shows potential relations when the linear operation is performed by using the pixel of Fig. 1A. In this case, a pulse R is applied to the gate of the first transistor 4. First, the gate of the first transistor 4 has been maintained at High level for a sufficiently long period before the start of signal storage, and the source potential VS comes to have a value VS(log) at which the photocurrent Ip and the subthreshold current Isubth become equal to each other. Next, the gate of the first transistor 4 changes to Low level, and the signal storage starts. After a lapse of one frame period, the source potential is reduced to VS(lin). Therefore, a variation ΔVS = VS(lin) - VS(log) of the source potential VS by the signal storage becomes: ΔVS = (Ip·ΔT)/C1 which expresses a value obtained by linearly converting the photocurrent Ip. In this case, ΔT represents the storage time, and C1 represents the capacitance of the detection node 3 shown in Fig. 1A.
    Fig. 1B illustrates the other Circuit construction example of each of the pixels 11 shown in Fig. 5 different from that of Fig. 1A. The circuit construction of Fig. 1B differs from the one of Fig. 1A in that a second transistor 2 is inserted between the photodiode 1 and the detection node 3. A DC potential T is applied to the gate of the transistor 2.
    Figs. 2B shows potential relations when the logarithmic operation is performed by using the pixel of Fig. 1B. Although a photocurrent Ip is generated in the photodiode 1, a current corresponding to the photocurrent Ip flows through the second transistor 2 in the steady state since the potential T at the gate of the second transistor 2 is a DC potential, and the potential of the photodiode 1 is maintained at a constant value. Further, in this case, the potential R of the gate of the first transistor 4 is a DC potential, and a subthreshold current Isubth flows. Since the potential VS of the detection node 3 changes so that the subthreshold current Isubth becomes equal to the photocurrent Ip, the potential VS of the detection node 3 comes to have a value VS (log) obtained by logarithmically converting the photocurrent Ip according to Equation (1). This makes it possible to achieve responses throughout a very wide range of the quantity of incident light and obtain a very wide dynamic range.
    Fig. 3B shows potential relations when the linear operation is performed by using the pixel of Fig. 1B. Although the photocurrent Ip is generated in the photodiode 1, a current corresponding to the photocurrent Ip flows through the gate in the steady state since the potential T of the gate of the second transistor 2 is a DC potential, and the potential of the photodiode 1 is maintained at a constant value. Further, in this case, the pulse R is applied to the gate of the first transistor 4. First, the first transistor 4 has been maintained at High level for a sufficiently long period before the signal storage starts, and the source voltage comes to have a value VS(log) at which the photocurrent Ip and the subthreshold current becomes equal to each other. Next, the gate potential of the first transistor 4 changes to Low level, and the signal storage starts. After a lapse of one frame, the source potential is reduced to VS(lin). Therefore, a variation ΔVS - VS(lin) - VS(log) of VS by the signal storage becomes: ΔVS = (Ip·ΔT)/C2 which expressed a value obtained by linearly converting the photocurrent Ip. In this case, ΔT represents the storage time, and C2 represents the capacitance of the detection node 3 shown in Fig. 1B. Although C1 = (capacitance of photodiode 1 + gate capacitance of transistor 6 + stray capacitance of wiring and so on), C2 = (gate capacitance of transistor 6 + stray capacitance of wiring and so on) and the area of the detection node 3 of Fig. 3B can be made sufficiently smaller than the area of the photodiode 1 of Fig. 3A, and therefore, C1 > C2. That is, it becomes possible to obtain in the case of Fig. 3B a signal voltage ΔVS higher than in the case of Fig. 3A with the same amount of signal charge IP·ΔT.
    Fig. 4A schematically shows the sectional structure of the pixel of Fig. 1A fabricated in a semiconductor substrate. Likewise, Figs. 4B and 4C schematically show the sectional structure of the pixel of Fig. 1B fabricated in a semiconductor substrate. A reference numeral 101 denotes a semiconductor substrate, 102 denotes a pixel isolation region, 103 denotes a cathode of the photodiode 1 (see Figs. 1A and 1B), 104 denotes the drain 5, and 111 denotes the first transistor 4. In Figs. 4B and 4C, a reference numeral 105 denotes an isolated detection node, which is separated from the cathode 103 of the photodiode 1 via the second transistor 112. Further, although the photodiode 1 has a simple PN junction structure and is formed concurrently with the drain 104 in Figs. 4A and 4B, the photodiode has a buried-channel structure and is formed separately from the drain in Fig. 4C. That is, a signal charge storage layer 106 is formed on the substrate side, and a heavily doped pinning layer 107 is formed on the surface side. In general, a buried-channel structure photodiode is allowed to have a largely reduced dark current in comparison with the simple PN junction structure. This makes it possible to extend the lower limit Imin of the photocurrent capable of being logarithmically converted during the logarithmic operation. Moreover, dark current noises can be reduced also in the linear operation.
    Fig. 6 shows the operation timing of the two-dimensional image sensor 10 shown in Fig. 5. In this case, R(1) and R(2) represent the reset clocks of the first row and the second row, S(1) and S(2) represent the pixel selection clocks of the first row and the second row, and OS represents an output signal. Moreover, 1H represents one horizontal scan period, and 1V represents one frame period. Paying attention to the pixels of the first row, first, in a frame as a preceding logarithmic operation period (shown at the left-hand end in Fig. 6), the reset clock, i.e., the gate potential R(1) of the first transistor 4 (see Figs. 1A and 1B) is maintained at High level to obtain a logarithmically converted photoelectric conversion signal at the detection node 3. Subsequently, the gate potential R(1) of the first transistor 4 is changed from High level to Low level for transition to the linear operation period. Then, by storing the photoelectrically converted electric charge in the detection node 3 only for one frame period, a linear type photoelectric conversion signal is obtained at the detection node 3. At this time, by turning on the pixel selection clock S(1) immediately before the gate potential R(1) changes from High level to Low level, a logarithmically converted photoelectric conversion signal Log(1) is outputted as the output signal OS. By turning on the pixel selection clock S(1) after one frame, i.e., immediately before the gate potential R(1) of the first transistor 4 changes from Low level to High level to transit from the linear operation period to the logarithmic operation period, a linear type photoelectric conversion signal Lin(1) is outputted as the output signal OS. The same thing can be said for the pixels of the second and subsequent rows except for sequential delays by one horizontal scan period. As described above, a frame in which the logarithmically converted photoelectric conversion signals Log(1), Log(2) and so on are obtained in units of 1H and a frame in which the linear type photoelectric conversion signals Lin(1), Lin(2) and so on are obtained in units of 1H alternate in the output signal OS.
    Figs. 7A and Fig. 7B are graphs showing a logarithmically converted photoelectric conversion signal Vs(log) and a linear type photoelectric conversion signal Vs(lin) obtained as the potentials of the detection node 3 of the pixel with the logarithm log(Ip) of the intensity of incident light represented by the horizontal axis (since the change in the potential Vs of the detection node 3 with respect to an increase in the incident light is in the negative direction in Figs. 1A and 1B, curves are inversely shown in Figs. 7A and 7B for the sake of convenience). In this case, the logarithmically converted photoelectric conversion signal Vs(log) does not depend on the read period, or the length of one frame period (properly referred to as an "1V period") in the case of Fig. 6. On the other hand, the linear type photoelectric conversion signal Vs(lin) has a signal storage period of 1V, and therefore, the output is increased as the period 1V is prolonged. Fig. 7A corresponds to the case where the period 1V is long, and Fig. 7B corresponds to the case where the period 1V is short. The value of Vs(log) with respect to the intensity of incident light has a response lower limit value Imin limited by the dark current and has an upper limit value being extremely higher than the linear type photoelectric conversion signal Vs(lin). On the other hand, the value of the linear type photoelectric conversion signal Vs(lin) with respect to the intensity of incident light has a form such that the linearly converted signal is superimposed on the value of the immediately preceding logarithmic conversion type photoelectric conversion signal Vs(log). Therefore, the net linear signal component becomes ΔVs = Vs (lin) - Vs(log). When the storage time is long as in Fig. 7A, the value of ΔVs becomes sufficiently greater than Vs(log), and Vs(lin) comes to have a linear graphic curve almost identical to ΔVs. When the storage time is short as in Fig. 7B, the value of ΔVs becomes lower than Vs(log) on the low incident light intensity side, and Vs(lin) comes to have a nonlinear graphic curve considerably changed from ΔVs.
    Fig. 7C is a graph showing the output signal of the image sensor of the present invention with the logarithm log(Ip) of the intensity of incident light represented by the horizontal axis. In this case, the solid line indicates a mean value <OS> of all the pixels, and the dashed line indicates a value OSij of a specified pixel (pixel having an address of i-th row and j-th column). In the case of the pixel shown in Fig. 1A, the response of each pixel accompanies a specific offset variation ΔVij due to the variations of the thresholds of the transistors 4 and 6, and the value of ΔVij varies every pixel. Therefore, when the response of each pixel is directly used as an image signal, the image quality is largely impaired by the harsh fixed pattern noises of ΔVij.
    Fig. 8A illustrates a circuit construction 30 for solving the problems concerning the fixed pattern noises. In the circuit construction 30, an analog signal from the image sensor 31 (identical to the image sensor 10 shown in Fig. 5) of the present invention is converted into a digital signal by an AD converter 33. The signal from the AD converter 33 is diverged so that one branch is guided directly to a difference circuit 37 as one example of a subtraction part and the other branch is guided to the difference circuit 37 via a frame memory 34 served as a first frame memory. When a logarithmically converted photoelectric conversion signal is outputted from the image sensor 31 under the condition that light is irradiated to each pixel with a uniform intensity (denoted by Ip1 in Fig. 7C), the signal is recorded in the frame memory 34 every pixel. By this operation, the offset variation ΔVij of each pixel is recorded in the frame memory 34. Next, the difference circuit 37 subtracts the signal recorded in the frame memory 34 from the signal read in an arbitrary frame under the subject imaging condition in association with each pixel. By this operation, the offset variation ΔVij is canceled in all the frame signals, i.e., the logarithmically converted photoelectric conversion signal OS(log) and the linear type photoelectric conversion signal OS(lin), and an image signal free from the fixed harsh noises can be obtained.
    The linear type photoelectric conversion signal OS(lin) in Fig. 8A has a characteristic that the logarithmic characteristic and the linear characteristic are added to each other although the offset variation ΔVij is canceled. For the above reasons, there is no problem because of an almost linear characteristic when the linear characteristic value is sufficiently greater than the logarithmic characteristic value as in Fig. 7A, a problem occurs due to a nonlinear characteristic when the linear characteristic value becomes lower than the logarithmic characteristic value as in Fig. 7B.
    Fig. 8B illustrates another circuit construction 40 for solving the problem concerning the linear characteristic. In the present circuit construction 40, an analog signal from the image sensor 31 of the present invention is converted into a digital signal by the AD converter 33. The signal from the AD converter 33 is diverged into three branches, and the first branch is directly guided to the difference circuit 37 as one example of the subtraction part. The second branch is guided to the difference circuit 37 via a frame memory 34 as a first frame memory and a changeover switch 36. Moreover, the third branch is guided to the difference circuit 37 via a frame memory 35 as a second frame memory and the changeover switch 36. A logarithmically converted photoelectric conversion signal from the image sensor 31 is recorded every pixel in the frame memory 34 under the condition that light is applied with a certain uniform intensity to each pixel as in the case of Fig. 8A. As a result, the offset variation ΔVij of each pixel is recorded in the frame memory 34. A logarithmic signal read immediately before the reset gate voltage changes from High level to Low level under the subject imaging condition is recorded in a rewrite manner in the frame memory 35. Since the changeover switch 36 is connected to the frame memory 35 side during linear signal read, the signal recorded in the frame memory 35 is subtracted from the linear signal read immediately before the reset gate voltage changes next from Low level to High level in correspondence with each pixel. As a result, only the net linear signal corresponding to the signal charge stored during the optical integration period is read. Furthermore, random noises accompanying the reset operation are also canceled through the subtraction process as a merit of the present technique, and therefore, not only the fixed noises but also the random noises can largely be reduced. On the other hand, the changeover switch 36 is connected to the frame memory 34 side during the logarithmic signal read, and therefore, the difference circuit 37 subtracts the signal recorded in the frame memory 34 from the logarithmic signal in correspondence with each pixel. As a result, the offset variation ΔVij is canceled in the logarithmically converted photoelectric conversion signal OS(log), and an image signal free from fixed harsh noises can be obtained.
    Although the case where the logarithmic operation period and the linear operation period are alternately repeated every one-frame period has been described in the example (Fig. 6), the present invention is not limited to this. Various combinations as shown in Figs. 9A, 9B and 9C are possible. Fig. 9A is the same as the case of Fig. 6, and Fig. 9B shows a case where the linear operation period of one frame and the logarithmic operation period of two frames are alternately repeated. Fig. 9C shows a case where the linear operation period of one frame and the logarithmic operation period of three frames are alternately repeated. Likewise, other combinations are, of course, possible.

    Claims (8)

    1. A solid-state imaging device in which a photodiode and a first transistor are provided in series between a ground and a drain in each pixel, and a signal corresponding to a current or an electric charge generated in the photodiode according to an optical input is outputted from a detection node located between the photodiode and the first transistor, comprising:
      a control part that executes control to alternately repeat a logarithmic operation period during which a photoelectric conversion signal logarithmically converted by setting a gate voltage of the first transistor to a first level is obtained and a linear operation period during which a linear photoelectric conversion signal is obtained by setting the gate voltage of the first transistor to a second level.
    2. The solid-state imaging device as claimed in claim 1, wherein
         the photodiode and the detection node are connected to each other.
    3. The solid-state imaging device as claimed in claim 1, wherein
         a second transistor is connected between the photodiode and the detection node.
    4. The solid-state imaging device as claimed in claim 3, wherein
         the photodiode has a buried-channel structure.
    5. The solid-state imaging device as claimed in claim 1, wherein
      the control part executes control so as to
      alternately repeat the logarithmic operation period and the linear operation period every frame,
      read a potential of the detection node as a linear type signal immediately before a transition from the linear operation period to the logarithmic operation period, and
      read the potential of the detection node as a logarithmic signal in the logarithmic operation period after a lapse of a certain period after the transition to the logarithmic operation period.
    6. The solid-state imaging device as claimed in claim 5, comprising:
      a first frame memory which stores a signal read from the detection node of each pixel in the logarithmic operation period under a condition that light is irradiated to each pixel with a certain uniform intensity; and
      a subtraction part which subsequently forms an output by subtracting the signal recorded in the first frame memory from a signal read in an arbitrary frame in association with each pixel.
    7. The solid-state imaging device as claimed in claim 6, wherein
         the subtraction part forms an output by subtracting the signal recorded in the first frame memory from a signal read in the logarithmic operation period in association with each pixel.
    8. The solid-state imaging device as claimed in claim 5, comprising:
      a second frame memory which records the signal read from the detection node every time immediately before the transition from the logarithmic operation period to the linear operation period under a subject imaging condition; and
      a subtraction part which subtracts the signal recorded in the second frame memory from the signal read from the detection node immediately before the transition from the linear operation period to the logarithmic operation period in association with each pixel.
    EP03768142A 2002-12-27 2003-12-24 SEMICONDUCTOR IMAGING DEVICE Withdrawn EP1580987A4 (en)

    Applications Claiming Priority (3)

    Application Number Priority Date Filing Date Title
    JP2002379240 2002-12-27
    JP2002379240A JP4185771B2 (en) 2002-12-27 2002-12-27 Solid-state imaging device
    PCT/JP2003/016550 WO2004062274A1 (en) 2002-12-27 2003-12-24 Solid-state imaging device

    Publications (2)

    Publication Number Publication Date
    EP1580987A1 true EP1580987A1 (en) 2005-09-28
    EP1580987A4 EP1580987A4 (en) 2007-10-24

    Family

    ID=32708384

    Family Applications (1)

    Application Number Title Priority Date Filing Date
    EP03768142A Withdrawn EP1580987A4 (en) 2002-12-27 2003-12-24 SEMICONDUCTOR IMAGING DEVICE

    Country Status (7)

    Country Link
    US (1) US7502060B2 (en)
    EP (1) EP1580987A4 (en)
    JP (1) JP4185771B2 (en)
    KR (1) KR100725764B1 (en)
    CN (1) CN100380933C (en)
    TW (1) TWI240570B (en)
    WO (1) WO2004062274A1 (en)

    Cited By (4)

    * Cited by examiner, † Cited by third party
    Publication number Priority date Publication date Assignee Title
    EP1648160A1 (en) * 2004-10-15 2006-04-19 OmniVision Technologies, Inc. Image sensor and pixel that has positive transfer gate voltage during integration period
    EP1845706A1 (en) * 2006-04-12 2007-10-17 CSEM Centre Suisse d'Electronique et de Microtechnique SA Photo sensor with pinned photodiode and sub-linear response
    US7378635B2 (en) 2005-02-11 2008-05-27 Micron Technology, Inc. Method and apparatus for dark current and hot pixel reduction in active pixel image sensors
    EP2352240A4 (en) * 2008-09-19 2014-05-14 Univ Shizuoka Nat Univ Corp INFORMATION ACQUISITION DEVICE AND OPTICAL COMMUNICATION SYSTEM

    Families Citing this family (24)

    * Cited by examiner, † Cited by third party
    Publication number Priority date Publication date Assignee Title
    US8913600B2 (en) 2003-01-07 2014-12-16 Sony Corporation Wireless communication apparatus, system, and device communicating with another wireless communication apparatus, system, and device in an autonomous distributed network without a designated control station and method thereof
    DE60333258D1 (en) * 2003-05-06 2010-08-19 St Microelectronics Res & Dev Combined linear and logarithmic image sensor
    JP4556722B2 (en) * 2004-05-31 2010-10-06 コニカミノルタホールディングス株式会社 Imaging device
    JP2005348005A (en) * 2004-06-02 2005-12-15 Konica Minolta Holdings Inc Imaging apparatus, imaging system, and imaging system operation program
    JP4581792B2 (en) 2004-07-05 2010-11-17 コニカミノルタホールディングス株式会社 Solid-state imaging device and camera equipped with the same
    JP4725052B2 (en) * 2004-08-17 2011-07-13 コニカミノルタホールディングス株式会社 Imaging device
    KR20060023890A (en) * 2004-09-11 2006-03-15 학교법인연세대학교 Broadband CMOS image sensor and optical response characteristics control method for controlling optical response characteristics
    WO2006085464A1 (en) * 2005-02-10 2006-08-17 Konica Minolta Holdings, Inc. Imaging device
    JP2007097127A (en) * 2005-08-30 2007-04-12 Konica Minolta Holdings Inc Solid-state imaging device
    JP4893042B2 (en) * 2006-03-20 2012-03-07 コニカミノルタホールディングス株式会社 Imaging device
    JP4873385B2 (en) * 2006-07-19 2012-02-08 オリンパス株式会社 Solid-state imaging device
    EP1940020A3 (en) * 2006-12-27 2009-03-04 Omron Corporation Solid-state imaging element, method of controlling solid-state imaging element, and imaging device
    JP2008263546A (en) * 2007-04-13 2008-10-30 Konica Minolta Holdings Inc Solid-state imaging apparatus, method for driving the solid-state imaging apparatus and imaging system using them
    JP5109684B2 (en) * 2007-06-22 2012-12-26 セイコーエプソン株式会社 Detection device and electronic device
    JP4927066B2 (en) 2007-12-26 2012-05-09 ローム アンド ハース カンパニー Curable composition
    JP4630907B2 (en) * 2008-03-03 2011-02-09 シャープ株式会社 Solid-state imaging device and electronic information device
    EP2579573A4 (en) * 2010-06-07 2014-09-10 Konica Minolta Advanced Layers IMAGING DEVICE
    EP2579574A4 (en) * 2010-06-07 2013-12-25 Konica Minolta Advanced Layers IMAGING DEVICE
    JP5178804B2 (en) * 2010-10-28 2013-04-10 ルネサスエレクトロニクス株式会社 Driving method of imaging apparatus
    JP2013058960A (en) * 2011-09-09 2013-03-28 Konica Minolta Advanced Layers Inc Solid-state imaging device
    JP6171997B2 (en) 2014-03-14 2017-08-02 ソニー株式会社 Solid-state imaging device, driving method thereof, and electronic apparatus
    JP6743011B2 (en) * 2014-12-11 2020-08-19 ソニーセミコンダクタソリューションズ株式会社 Imaging device, driving method, electronic device
    EP3955566B1 (en) * 2020-08-14 2023-06-07 Alpsentek GmbH Image sensor with configurable pixel circuit and method
    TWI828118B (en) * 2022-04-19 2024-01-01 睿生光電股份有限公司 Detection device

    Family Cites Families (13)

    * Cited by examiner, † Cited by third party
    Publication number Priority date Publication date Assignee Title
    US5289286A (en) * 1991-07-18 1994-02-22 Minolta Camera Kabushiki Kaisha Solid state sensor having logarithmic photovoltaic response, with pixel uniformity correction and white balance circuitry therefor
    JPH0530350A (en) * 1991-07-18 1993-02-05 Minolta Camera Co Ltd Solid-state image pickup device
    US5471515A (en) * 1994-01-28 1995-11-28 California Institute Of Technology Active pixel sensor with intra-pixel charge transfer
    JP3576715B2 (en) 1996-09-10 2004-10-13 本田技研工業株式会社 Optical sensor circuit
    JP2000175108A (en) 1998-12-04 2000-06-23 Honda Motor Co Ltd Output correction circuit of image sensor
    JP3664035B2 (en) * 1999-03-29 2005-06-22 コニカミノルタホールディングス株式会社 Solid-state imaging device
    US7030921B2 (en) * 2000-02-01 2006-04-18 Minolta Co., Ltd. Solid-state image-sensing device
    JP3856283B2 (en) * 2000-02-14 2006-12-13 シャープ株式会社 Solid-state imaging device and driving method of imaging device
    US20010015404A1 (en) * 2000-02-18 2001-08-23 Minolta Co., Ltd. Solid-state image-sensing device
    JP4306090B2 (en) * 2000-05-25 2009-07-29 日本ビクター株式会社 Solid-state imaging device
    JP4374745B2 (en) * 2000-07-19 2009-12-02 コニカミノルタホールディングス株式会社 Solid-state imaging device
    EP1265291A1 (en) * 2001-06-08 2002-12-11 EM Microelectronic-Marin SA CMOS image sensor and method for driving a CMOS image sensor with increased dynamic range
    US7012238B2 (en) * 2003-07-02 2006-03-14 Sharp Kabushiki Kaisha Amplification-type solid-state image pickup device incorporating plurality of arrayed pixels with amplification function

    Cited By (9)

    * Cited by examiner, † Cited by third party
    Publication number Priority date Publication date Assignee Title
    EP1648160A1 (en) * 2004-10-15 2006-04-19 OmniVision Technologies, Inc. Image sensor and pixel that has positive transfer gate voltage during integration period
    CN100399806C (en) * 2004-10-15 2008-07-02 豪威科技有限公司 Image sensor and pixel with positive voltage on transfer gate during integration period
    US7791663B2 (en) 2004-10-15 2010-09-07 Omnivision Technologies, Inc. Image sensor and pixel that has positive transfer gate voltage during integration period
    US7956913B2 (en) 2004-10-15 2011-06-07 Omnivision Technologies, Inc. Image sensor and pixel that has positive transfer gate voltage during integration period
    US7378635B2 (en) 2005-02-11 2008-05-27 Micron Technology, Inc. Method and apparatus for dark current and hot pixel reduction in active pixel image sensors
    EP1845706A1 (en) * 2006-04-12 2007-10-17 CSEM Centre Suisse d'Electronique et de Microtechnique SA Photo sensor with pinned photodiode and sub-linear response
    WO2007115415A1 (en) * 2006-04-12 2007-10-18 Csem Centre Suisse D'electronique Et De Microtechnique Sa Photo sensor with pinned photodiode and sub-linear response
    EP2352240A4 (en) * 2008-09-19 2014-05-14 Univ Shizuoka Nat Univ Corp INFORMATION ACQUISITION DEVICE AND OPTICAL COMMUNICATION SYSTEM
    US8891978B2 (en) 2008-09-19 2014-11-18 National University Corporation Shizuoka University Information-acquisition device and optical communication system

    Also Published As

    Publication number Publication date
    US7502060B2 (en) 2009-03-10
    TWI240570B (en) 2005-09-21
    TW200425729A (en) 2004-11-16
    US20060044436A1 (en) 2006-03-02
    EP1580987A4 (en) 2007-10-24
    KR20050087856A (en) 2005-08-31
    CN1757227A (en) 2006-04-05
    WO2004062274A1 (en) 2004-07-22
    JP4185771B2 (en) 2008-11-26
    KR100725764B1 (en) 2007-06-08
    JP2004214772A (en) 2004-07-29
    CN100380933C (en) 2008-04-09

    Similar Documents

    Publication Publication Date Title
    US7502060B2 (en) Solid-state imaging device providing wide dynamic range and high low-illuminance sensitivity
    US10567691B2 (en) Solid-state imaging device, method for driving solid-state imaging device, and electronic apparatus
    US7791657B2 (en) Dynamic range enhancement scheme for imagers
    KR101437912B1 (en) How to drive the image sensor
    JP5012188B2 (en) Solid-state imaging device
    JP2009177749A (en) Solid-state imaging device
    JP3927696B2 (en) Imaging device
    TW200838288A (en) Image sensors with output noise reduction mechanisms
    US6831691B1 (en) Solid-state image pickup device
    US6995797B2 (en) Charge detecting device for a solid state imaging device
    JP4956750B2 (en) Pixel for image sensor and image sensor device
    JP4770618B2 (en) Solid-state imaging device
    US8045028B1 (en) Six transistor (6T) pixel architecture
    JP4300654B2 (en) Solid-state imaging device
    US7012238B2 (en) Amplification-type solid-state image pickup device incorporating plurality of arrayed pixels with amplification function
    JP2000209508A (en) Solid-state imaging device
    JPWO2007066762A1 (en) Solid-state imaging device
    YusukeOike et al. A 121.8 dB dynamic range CMOS image sensor using pixel-variation-free midpoint potential drive and overlapping multiple exposures
    JP2016178408A (en) Solid-state image pickup device, driving method for the same and imaging system
    JP4345145B2 (en) Solid-state imaging device
    JP2001036059A (en) Solid-stage image pickup device
    JPH1093868A (en) Solid-state imaging device and driving method thereof
    JP3500761B2 (en) Solid-state imaging device and driving method thereof
    US20250280211A1 (en) Imaging device
    JP2001218111A (en) Solid-state image pickup device

    Legal Events

    Date Code Title Description
    PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

    Free format text: ORIGINAL CODE: 0009012

    17P Request for examination filed

    Effective date: 20050720

    AK Designated contracting states

    Kind code of ref document: A1

    Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LI LU MC NL PT RO SE SI SK TR

    AX Request for extension of the european patent

    Extension state: AL LT LV MK

    DAX Request for extension of the european patent (deleted)
    RBV Designated contracting states (corrected)

    Designated state(s): DE FR IT

    A4 Supplementary search report drawn up and despatched

    Effective date: 20070921

    STAA Information on the status of an ep patent application or granted ep patent

    Free format text: STATUS: THE APPLICATION HAS BEEN WITHDRAWN

    18W Application withdrawn

    Effective date: 20080814